The goal of this application is to elucidate dietary factors that elevate risk for cardiovascular disease (CVD) in conjunction with insulin resistance. Potential dietary determinants of hyperinsulinemia or of insulin resistance will be evaluated, including high intake of dietary fats and of simple sugars and low alcohol consumption. As a major focus, dietary factors that may contribute to elevated CVD risk among persons with existing insulin resistance, including impaired glucose tolerance (IGT) and non-insulin- dependent diabetes mellitus (NIDDM) will be examined. Clinical studies indicate that high carbohydrate intake may exacerbate the dyslipidemia often seen with insulin resistance. Consequently, there is much debate as to the optimal dietary pattern for individuals with NIDDM to minimize risk for microvascular disease. Data on this topic from free-living populations are scarce. In addition, associations of dietary antioxidants with CVD risk have not been evaluated in large community samples of persons with diabetes. Five datasets from epidemiologic studies will be available to evaluate specific hypotheses for men and women of black, Hispanic and non-Hispanic white ethnicity. The unique contributions of each dataset are as follow. The San Luis Valley Diabetes Study and the San Antonio Heart Study enable prospective analyses for large numbers of subjects with IGT or NIDDM; the Mexico City Study provides a sample with contrasting dietary patterns in a non-white population; the Kaiser Permanente Women Twins Study allows for the removal of genetic influences by evaluating associations within monozygotic twin pairs; and the Insulin Resistance and Atherosclerosis Study (IRAS) provides direct measurement of insulin sensitivity and subclinical atherosclerosis across the spectrum of glucose tolerance in three ethnic groups. An understanding of the accuracy of the dietary assessment instruments used in the three ethnic groups is critical to the interpretation of the findings related to diet and CVD risk variables. Therefore, a second component of research is also proposed, that being an evaluation of the comparative validity of the food frequency interview used in the multi- cultural IRAS population using a series of 24-hour dietary recalls as the standard.